Auto merge of #78296 - Aaron1011:fix/stmt-tokens, r=petrochenkov

Properly handle attributes on statements

We now collect tokens for the underlying node wrapped by `StmtKind`
nstead of storing tokens directly in `Stmt`.

`LazyTokenStream` now supports capturing a trailing semicolon after it
is initially constructed. This allows us to avoid refactoring statement
parsing to wrap the parsing of the semicolon in `parse_tokens`.

Attributes on item statements
(e.g. `fn foo() { #[bar] struct MyStruct; }`) are now treated as
item attributes, not statement attributes, which is consistent with how
we handle attributes on other kinds of statements. The feature-gating
code is adjusted so that proc-macro attributes are still allowed on item
statements on stable.

Two built-in macros (`#[global_allocator]` and `#[test]`) needed to be
adjusted to support being passed `Annotatable::Stmt`.

compiler/rustc_ast/src/ast.rs

@@ -901,10 +901,39 @@ pub struct Stmt {
     pub id: NodeId,
     pub kind: StmtKind,
     pub span: Span,
-    pub tokens: Option<LazyTokenStream>,
 }
 
 impl Stmt {
+    pub fn tokens(&self) -> Option<&LazyTokenStream> {
+        match self.kind {
+            StmtKind::Local(ref local) => local.tokens.as_ref(),
+            StmtKind::Item(ref item) => item.tokens.as_ref(),
+            StmtKind::Expr(ref expr) | StmtKind::Semi(ref expr) => expr.tokens.as_ref(),
+            StmtKind::Empty => None,
+            StmtKind::MacCall(ref mac) => mac.tokens.as_ref(),
+        }
+    }
+
+    pub fn tokens_mut(&mut self) -> Option<&mut LazyTokenStream> {
+        match self.kind {
+            StmtKind::Local(ref mut local) => local.tokens.as_mut(),
+            StmtKind::Item(ref mut item) => item.tokens.as_mut(),
+            StmtKind::Expr(ref mut expr) | StmtKind::Semi(ref mut expr) => expr.tokens.as_mut(),
+            StmtKind::Empty => None,
+            StmtKind::MacCall(ref mut mac) => mac.tokens.as_mut(),
+        }
+    }
+
+    pub fn set_tokens(&mut self, tokens: Option<LazyTokenStream>) {
+        match self.kind {
+            StmtKind::Local(ref mut local) => local.tokens = tokens,
+            StmtKind::Item(ref mut item) => item.tokens = tokens,
+            StmtKind::Expr(ref mut expr) | StmtKind::Semi(ref mut expr) => expr.tokens = tokens,
+            StmtKind::Empty => {}
+            StmtKind::MacCall(ref mut mac) => mac.tokens = tokens,
+        }
+    }
+
     pub fn has_trailing_semicolon(&self) -> bool {
         match &self.kind {
             StmtKind::Semi(_) => true,
@@ -912,18 +941,25 @@ impl Stmt {
             _ => false,
         }
     }
+
+    /// Converts a parsed `Stmt` to a `Stmt` with
+    /// a trailing semicolon.
+    ///
+    /// This only modifies the parsed AST struct, not the attached
+    /// `LazyTokenStream`. The parser is responsible for calling
+    /// `CreateTokenStream::add_trailing_semi` when there is actually
+    /// a semicolon in the tokenstream.
     pub fn add_trailing_semicolon(mut self) -> Self {
         self.kind = match self.kind {
             StmtKind::Expr(expr) => StmtKind::Semi(expr),
             StmtKind::MacCall(mac) => {
-                StmtKind::MacCall(mac.map(|MacCallStmt { mac, style: _, attrs }| MacCallStmt {
-                    mac,
-                    style: MacStmtStyle::Semicolon,
-                    attrs,
+                StmtKind::MacCall(mac.map(|MacCallStmt { mac, style: _, attrs, tokens }| {
+                    MacCallStmt { mac, style: MacStmtStyle::Semicolon, attrs, tokens }
                 }))
             }
             kind => kind,
         };
+
         self
     }
 
@@ -963,6 +999,7 @@ pub struct MacCallStmt {
     pub mac: MacCall,
     pub style: MacStmtStyle,
     pub attrs: AttrVec,
+    pub tokens: Option<LazyTokenStream>,
 }
 
 #[derive(Clone, Copy, PartialEq, Encodable, Decodable, Debug)]
@@ -988,6 +1025,7 @@ pub struct Local {
     pub init: Option<P<Expr>>,
     pub span: Span,
     pub attrs: AttrVec,
+    pub tokens: Option<LazyTokenStream>,
 }
 
 /// An arm of a 'match'.

compiler/rustc_ast/src/mut_visit.rs

@@ -579,13 +579,14 @@ pub fn noop_visit_parenthesized_parameter_data<T: MutVisitor>(
 }
 
 pub fn noop_visit_local<T: MutVisitor>(local: &mut P<Local>, vis: &mut T) {
-    let Local { id, pat, ty, init, span, attrs } = local.deref_mut();
+    let Local { id, pat, ty, init, span, attrs, tokens } = local.deref_mut();
     vis.visit_id(id);
     vis.visit_pat(pat);
     visit_opt(ty, |ty| vis.visit_ty(ty));
     visit_opt(init, |init| vis.visit_expr(init));
     vis.visit_span(span);
     visit_thin_attrs(attrs, vis);
+    visit_lazy_tts(tokens, vis);
 }
 
 pub fn noop_visit_attribute<T: MutVisitor>(attr: &mut Attribute, vis: &mut T) {
@@ -1328,16 +1329,12 @@ pub fn noop_filter_map_expr<T: MutVisitor>(mut e: P<Expr>, vis: &mut T) -> Optio
 }
 
 pub fn noop_flat_map_stmt<T: MutVisitor>(
-    Stmt { kind, mut span, mut id, mut tokens }: Stmt,
+    Stmt { kind, mut span, mut id }: Stmt,
     vis: &mut T,
 ) -> SmallVec<[Stmt; 1]> {
     vis.visit_id(&mut id);
     vis.visit_span(&mut span);
-    visit_lazy_tts(&mut tokens, vis);
-    noop_flat_map_stmt_kind(kind, vis)
-        .into_iter()
-        .map(|kind| Stmt { id, kind, span, tokens: tokens.clone() })
-        .collect()
+    noop_flat_map_stmt_kind(kind, vis).into_iter().map(|kind| Stmt { id, kind, span }).collect()
 }
 
 pub fn noop_flat_map_stmt_kind<T: MutVisitor>(
@@ -1354,9 +1351,10 @@ pub fn noop_flat_map_stmt_kind<T: MutVisitor>(
         StmtKind::Semi(expr) => vis.filter_map_expr(expr).into_iter().map(StmtKind::Semi).collect(),
         StmtKind::Empty => smallvec![StmtKind::Empty],
         StmtKind::MacCall(mut mac) => {
-            let MacCallStmt { mac: mac_, style: _, attrs } = mac.deref_mut();
+            let MacCallStmt { mac: mac_, style: _, attrs, tokens } = mac.deref_mut();
             vis.visit_mac_call(mac_);
             visit_thin_attrs(attrs, vis);
+            visit_lazy_tts(tokens, vis);
             smallvec![StmtKind::MacCall(mac)]
         }
     }

compiler/rustc_ast/src/tokenstream.rs

@@ -121,10 +121,14 @@ where
 }
 
 pub trait CreateTokenStream: sync::Send + sync::Sync {
+    fn add_trailing_semi(&self) -> Box<dyn CreateTokenStream>;
     fn create_token_stream(&self) -> TokenStream;
 }
 
 impl CreateTokenStream for TokenStream {
+    fn add_trailing_semi(&self) -> Box<dyn CreateTokenStream> {
+        panic!("Cannot call `add_trailing_semi` on a `TokenStream`!");
+    }
     fn create_token_stream(&self) -> TokenStream {
         self.clone()
     }
@@ -141,6 +145,13 @@ impl LazyTokenStream {
         LazyTokenStream(Lrc::new(Box::new(inner)))
     }
 
+    /// Extends the captured stream by one token,
+    /// which must be a trailing semicolon. This
+    /// affects the `TokenStream` created by `make_tokenstream`.
+    pub fn add_trailing_semi(&self) -> LazyTokenStream {
+        LazyTokenStream(Lrc::new(self.0.add_trailing_semi()))
+    }
+
     pub fn create_token_stream(&self) -> TokenStream {
         self.0.create_token_stream()
     }

compiler/rustc_ast/src/visit.rs

@@ -689,7 +689,7 @@ pub fn walk_stmt<'a, V: Visitor<'a>>(visitor: &mut V, statement: &'a Stmt) {
         StmtKind::Expr(ref expr) | StmtKind::Semi(ref expr) => visitor.visit_expr(expr),
         StmtKind::Empty => {}
         StmtKind::MacCall(ref mac) => {
-            let MacCallStmt { ref mac, style: _, ref attrs } = **mac;
+            let MacCallStmt { ref mac, style: _, ref attrs, tokens: _ } = **mac;
             visitor.visit_mac_call(mac);
             for attr in attrs.iter() {
                 visitor.visit_attribute(attr);